Honors Theses
Date of Award
Spring 5-8-2026
Document Type
Undergraduate Thesis
Department
Computer and Information Science
First Advisor
Bo Wang
Second Advisor
Timothy Holston
Relational Format
Dissertation/Thesis
Abstract
Running gait analysis plays a critical role in injury prevention and performance optimization, however, existing approaches often rely on specialized laboratory equipment or wearable sensors with limited interpretability. Recent advances in computer vision, particularly 2D human pose estimation, enable markerless motion analysis from standard video. However, progress remains constrained by the lack of publicly available datasets designed for running form analysis.
In this work, we introduce a preliminary dataset and benchmark for stride-level running gait analysis. The dataset consists of 73 treadmill running videos from 15 participants with varying experience levels, annotated with over 4,600 stride-level labels across multiple biomechanical cues. We define a set of running form deviations and extract interpretable stride-level features grounded in biomechanical principles. Using these features, we formulate running form assessment as a multi-label classification task and evaluate several baseline models.
Our results show that the proposed feature representation effectively detects several running form deviations, with XGBoost achieving the best overall performance. At the same time, challenges remain in detecting more complex cues such as excessive vertical oscillation, highlighting limitations of 2D pose-based representations. This work provides an initial benchmark for video-based running gait analysis and lays the foundation for future research in data-driven assessment of running form.
Recommended Citation
Eguibar Ortega, Paulina, "Stridevision: Automated Detection of Running Form Deviations from 2D Pose Estimation and Machine Learning" (2026). Honors Theses. 3469.
https://egrove.olemiss.edu/hon_thesis/3469